In [7]:
# view histograms for numerical features
for column in num_features:
desc_num_feature(column)
# import libraries
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
import os
import pandas_profiling
from sklearn.preprocessing import MinMaxScaler
%matplotlib inline
# import dataset
insurance= pd.read_csv('insurance.csv')
# view dataset
insurance.sample(10)
| age | sex | bmi | children | smoker | region | charges | |
|---|---|---|---|---|---|---|---|
| 1011 | 43 | male | 20.130 | 2 | yes | southeast | 18767.73770 |
| 131 | 61 | female | 22.040 | 0 | no | northeast | 13616.35860 |
| 105 | 20 | male | 28.025 | 1 | yes | northwest | 17560.37975 |
| 816 | 24 | female | 24.225 | 0 | no | northwest | 2842.76075 |
| 679 | 49 | female | 23.180 | 2 | no | northwest | 10156.78320 |
| 1138 | 33 | male | 30.250 | 0 | no | southeast | 3704.35450 |
| 1064 | 29 | female | 25.600 | 4 | no | southwest | 5708.86700 |
| 564 | 18 | female | 32.120 | 2 | no | southeast | 2801.25880 |
| 578 | 52 | male | 30.200 | 1 | no | southwest | 9724.53000 |
| 1218 | 46 | female | 34.600 | 1 | yes | southwest | 41661.60200 |
# information on dataset
insurance.info()
<class 'pandas.core.frame.DataFrame'> RangeIndex: 1338 entries, 0 to 1337 Data columns (total 7 columns): # Column Non-Null Count Dtype --- ------ -------------- ----- 0 age 1338 non-null int64 1 sex 1338 non-null object 2 bmi 1338 non-null float64 3 children 1338 non-null int64 4 smoker 1338 non-null object 5 region 1338 non-null object 6 charges 1338 non-null float64 dtypes: float64(2), int64(2), object(3) memory usage: 73.3+ KB
# categorize features
num_features= ['age', 'bmi', 'charges']
cat_features= ['sex', 'children', 'smoker', 'region']
# numerical features histograms
'''
Hands On Predictive Analytics with Python
Alvaro Fuentes
'''
def desc_num_feature(feature_name, bins= 30, edgecolor= 'k', **kwargs):
fig, ax= plt.subplots(figsize= (8, 4))
insurance[feature_name].hist(bins= bins, edgecolor= edgecolor, ax= ax, **kwargs)
ax.set_title(feature_name, size= 15)
plt.figtext(1, 0.15, str(insurance[feature_name].describe().round(2)), size= 17)
# view histograms for numerical features
for column in num_features:
desc_num_feature(column)
# categorical feature bar graphs
'''
Hands On Predictive Analytics with Python
Alvaro Fuentes
'''
def cat_feature_bar_chart(feature):
count= insurance[feature].value_counts()
percent= 100 * insurance[feature].value_counts(normalize= True)
df= pd.DataFrame({'count': count, 'percent': percent.round(1)})
print(df)
count.plot(kind= 'bar', title= feature)
# view categoical bar charts
cat_feature_bar_chart('sex')
count percent male 676 50.5 female 662 49.5
cat_feature_bar_chart('children')
count percent 0 574 42.9 1 324 24.2 2 240 17.9 3 157 11.7 4 25 1.9 5 18 1.3
cat_feature_bar_chart('smoker')
count percent no 1064 79.5 yes 274 20.5
cat_feature_bar_chart('region')
count percent southeast 364 27.2 southwest 325 24.3 northwest 325 24.3 northeast 324 24.2
# Profiling report
pandas_profiling.ProfileReport(insurance)
Summarize dataset: 0%| | 0/5 [00:00<?, ?it/s]
Generate report structure: 0%| | 0/1 [00:00<?, ?it/s]
Render HTML: 0%| | 0/1 [00:00<?, ?it/s]
# one hot encoding for categorical values
sex= pd.get_dummies(insurance.sex, prefix= 'sex')
smoker= pd.get_dummies(insurance.smoker, prefix= 'smoker')
region= pd.get_dummies(insurance.region, prefix= 'region')
# combine datasets
insurance = pd.concat([insurance, sex, smoker, region], axis= 1)
# view data
insurance.sample(10)
| age | sex | bmi | children | smoker | region | charges | sex_female | sex_male | smoker_no | smoker_yes | region_northeast | region_northwest | region_southeast | region_southwest | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 674 | 44 | female | 43.890 | 2 | yes | southeast | 46200.98510 | 1 | 0 | 0 | 1 | 0 | 0 | 1 | 0 |
| 166 | 20 | female | 37.000 | 5 | no | southwest | 4830.63000 | 1 | 0 | 1 | 0 | 0 | 0 | 0 | 1 |
| 493 | 61 | male | 43.400 | 0 | no | southwest | 12574.04900 | 0 | 1 | 1 | 0 | 0 | 0 | 0 | 1 |
| 1187 | 62 | female | 32.680 | 0 | no | northwest | 13844.79720 | 1 | 0 | 1 | 0 | 0 | 1 | 0 | 0 |
| 414 | 19 | female | 35.150 | 0 | no | northwest | 2134.90150 | 1 | 0 | 1 | 0 | 0 | 1 | 0 | 0 |
| 335 | 64 | male | 34.500 | 0 | no | southwest | 13822.80300 | 0 | 1 | 1 | 0 | 0 | 0 | 0 | 1 |
| 66 | 61 | female | 39.100 | 2 | no | southwest | 14235.07200 | 1 | 0 | 1 | 0 | 0 | 0 | 0 | 1 |
| 905 | 26 | female | 29.355 | 2 | no | northeast | 4564.19145 | 1 | 0 | 1 | 0 | 1 | 0 | 0 | 0 |
| 176 | 38 | male | 27.835 | 2 | no | northwest | 6455.86265 | 0 | 1 | 1 | 0 | 0 | 1 | 0 | 0 |
| 241 | 33 | female | 22.135 | 1 | no | northeast | 5354.07465 | 1 | 0 | 1 | 0 | 1 | 0 | 0 | 0 |
# drop unnedded columns
insurance= insurance.drop('sex', axis= 1)
insurance= insurance.drop('smoker', axis= 1)
insurance= insurance.drop('region', axis= 1)
# view finished dataset
insurance.sample(10)
| age | bmi | children | charges | sex_female | sex_male | smoker_no | smoker_yes | region_northeast | region_northwest | region_southeast | region_southwest | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 32 | 19 | 28.600 | 5 | 4687.79700 | 1 | 0 | 1 | 0 | 0 | 0 | 0 | 1 |
| 754 | 24 | 33.630 | 4 | 17128.42608 | 0 | 1 | 1 | 0 | 1 | 0 | 0 | 0 |
| 865 | 40 | 29.900 | 2 | 6600.36100 | 0 | 1 | 1 | 0 | 0 | 0 | 0 | 1 |
| 140 | 34 | 22.420 | 2 | 27375.90478 | 0 | 1 | 1 | 0 | 1 | 0 | 0 | 0 |
| 412 | 26 | 17.195 | 2 | 14455.64405 | 1 | 0 | 0 | 1 | 1 | 0 | 0 | 0 |
| 985 | 44 | 25.800 | 1 | 7624.63000 | 1 | 0 | 1 | 0 | 0 | 0 | 0 | 1 |
| 35 | 19 | 20.425 | 0 | 1625.43375 | 0 | 1 | 1 | 0 | 0 | 1 | 0 | 0 |
| 290 | 28 | 33.400 | 0 | 3172.01800 | 1 | 0 | 1 | 0 | 0 | 0 | 0 | 1 |
| 566 | 38 | 40.565 | 1 | 6373.55735 | 1 | 0 | 1 | 0 | 0 | 1 | 0 | 0 |
| 67 | 40 | 26.315 | 1 | 6389.37785 | 0 | 1 | 1 | 0 | 0 | 1 | 0 | 0 |
# min max scaling of numeric columns
num_features= ['age', 'bmi', 'children']
scaler= MinMaxScaler()
scaler.fit(insurance[num_features])
MinMaxScaler()
insurance.loc[:, num_features]= scaler.transform(insurance[num_features])
insurance.sample(10)
| age | bmi | children | charges | sex_female | sex_male | smoker_no | smoker_yes | region_northeast | region_northwest | region_southeast | region_southwest | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 393 | 0.673913 | 0.414044 | 0.2 | 9290.13950 | 0 | 1 | 1 | 0 | 1 | 0 | 0 | 0 |
| 579 | 0.152174 | 0.201910 | 0.0 | 3206.49135 | 1 | 0 | 1 | 0 | 1 | 0 | 0 | 0 |
| 477 | 0.152174 | 0.529056 | 0.0 | 2534.39375 | 0 | 1 | 1 | 0 | 0 | 1 | 0 | 0 |
| 1270 | 0.173913 | 0.483051 | 0.2 | 3292.52985 | 0 | 1 | 1 | 0 | 0 | 1 | 0 | 0 |
| 767 | 0.521739 | 0.350821 | 0.2 | 7050.64200 | 1 | 0 | 1 | 0 | 0 | 0 | 0 | 1 |
| 1210 | 0.391304 | 0.401264 | 0.2 | 5373.36425 | 0 | 1 | 1 | 0 | 0 | 1 | 0 | 0 |
| 626 | 0.391304 | 0.347592 | 0.6 | 6748.59120 | 0 | 1 | 1 | 0 | 1 | 0 | 0 | 0 |
| 1208 | 0.413043 | 0.258138 | 0.2 | 20296.86345 | 1 | 0 | 0 | 1 | 1 | 0 | 0 | 0 |
| 1219 | 0.434783 | 0.383374 | 0.6 | 7537.16390 | 1 | 0 | 1 | 0 | 0 | 1 | 0 | 0 |
| 506 | 0.086957 | 0.414044 | 0.2 | 2643.26850 | 0 | 1 | 1 | 0 | 0 | 1 | 0 | 0 |
insurance.describe()
| age | bmi | children | charges | sex_female | sex_male | smoker_no | smoker_yes | region_northeast | region_northwest | region_southeast | region_southwest | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| count | 1338.000000 | 1338.000000 | 1338.000000 | 1338.000000 | 1338.000000 | 1338.000000 | 1338.000000 | 1338.000000 | 1338.000000 | 1338.000000 | 1338.000000 | 1338.000000 |
| mean | 0.461022 | 0.395572 | 0.218984 | 13270.422265 | 0.494768 | 0.505232 | 0.795217 | 0.204783 | 0.242152 | 0.242900 | 0.272048 | 0.242900 |
| std | 0.305434 | 0.164062 | 0.241099 | 12110.011237 | 0.500160 | 0.500160 | 0.403694 | 0.403694 | 0.428546 | 0.428995 | 0.445181 | 0.428995 |
| min | 0.000000 | 0.000000 | 0.000000 | 1121.873900 | 0.000000 | 0.000000 | 0.000000 | 0.000000 | 0.000000 | 0.000000 | 0.000000 | 0.000000 |
| 25% | 0.195652 | 0.278080 | 0.000000 | 4740.287150 | 0.000000 | 0.000000 | 1.000000 | 0.000000 | 0.000000 | 0.000000 | 0.000000 | 0.000000 |
| 50% | 0.456522 | 0.388485 | 0.200000 | 9382.033000 | 0.000000 | 1.000000 | 1.000000 | 0.000000 | 0.000000 | 0.000000 | 0.000000 | 0.000000 |
| 75% | 0.717391 | 0.504002 | 0.400000 | 16639.912515 | 1.000000 | 1.000000 | 1.000000 | 0.000000 | 0.000000 | 0.000000 | 1.000000 | 0.000000 |
| max | 1.000000 | 1.000000 | 1.000000 | 63770.428010 | 1.000000 | 1.000000 | 1.000000 | 1.000000 | 1.000000 | 1.000000 | 1.000000 | 1.000000 |
from sklearn import linear_model
from sklearn.model_selection import train_test_split
from sklearn.metrics import r2_score, mean_squared_error, mean_absolute_error, mean_absolute_percentage_error
# Set my predictor field and my features
X = insurance.drop(['charges'], axis=1)
y = insurance.charges
# Train-test-split function to retain 40% of the data for testing
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.4, random_state=0)
# Instantiate the model and fit the training data
reg = linear_model.LinearRegression()
reg.fit(X_train, y_train)
LinearRegression()
# Calculate weights or coefficients of the linear model
pd.Series(reg.coef_, index=X_train.columns).sort_values(ascending=False)
smoker_yes 7.251810e+16 smoker_no 7.251810e+16 bmi 1.268353e+04 age 1.162991e+04 children 2.596428e+03 sex_male -1.434257e+16 sex_female -1.434257e+16 region_northwest -5.276846e+17 region_northeast -5.276846e+17 region_southwest -5.276846e+17 region_southeast -5.276846e+17 dtype: float64
# Predict on the train data and get the accuracy
reg_predict_train = reg.predict(X_train)
coef_of_determine_train = r2_score(y_train, reg_predict_train)
print("Training Coefficient of Determination is {:.2%}".format(coef_of_determine_train))
Training Coefficient of Determination is 72.15%
# Predict on the test data and get the accuracy
reg_predict_test = reg.predict(X_test)
coef_of_determine_test = r2_score(y_test, reg_predict_test)
print("Testing Coefficient of Determination is {:.2%}".format(coef_of_determine_test))
Testing Coefficient of Determination is 78.72%
# Find MSE for training data
mse = mean_squared_error(y_true=y_test, y_pred=reg_predict_test)
print("Mean Square Error is", round(mse, 0))
Mean Square Error is 32940508.0
mae = mean_absolute_error(y_true=y_test, y_pred=reg_predict_test)
print("Mean Absolute Error is", round(mae, 0))
Mean Absolute Error is 4061.0
mape = mean_absolute_percentage_error(y_true=y_test, y_pred=reg_predict_test)
print("Mean Absolute Percentage Error is {:.2%}".format(mape))
Mean Absolute Percentage Error is 42.46%
df = pd.read_csv('insurance.csv')
pandas_profiling.ProfileReport(df)
Summarize dataset: 0%| | 0/5 [00:00<?, ?it/s]
Generate report structure: 0%| | 0/1 [00:00<?, ?it/s]
Render HTML: 0%| | 0/1 [00:00<?, ?it/s]
plt.figure(figsize = (10, 10))
sns.heatmap(df.corr())
<AxesSubplot:>
new = df.corr()
new
| age | bmi | children | charges | |
|---|---|---|---|---|
| age | 1.000000 | 0.109272 | 0.042469 | 0.299008 |
| bmi | 0.109272 | 1.000000 | 0.012759 | 0.198341 |
| children | 0.042469 | 0.012759 | 1.000000 | 0.067998 |
| charges | 0.299008 | 0.198341 | 0.067998 | 1.000000 |
f = plt.figure()
f.set_figwidth(15)
f.set_figheight(10)
plt.scatter(df.age, df.charges)
plt.title('Charges by Age')
plt.xlabel('Age')
plt.ylabel('Charges')
plt.show()
plt.bar(df.smoker, df.charges)
plt.title('Charges by Smoker Category')
plt.xlabel('Smoker')
plt.ylabel('Charges')
plt.show()
f = plt.figure()
f.set_figwidth(15)
f.set_figheight(10)
plt.scatter(df.bmi, df.charges, color = 'blue')
plt.title('Charges by BMI')
plt.xlabel('BMI')
plt.ylabel('Charges')
plt.show()
f = plt.figure()
f.set_figwidth(15)
f.set_figheight(10)
plt.scatter(df.children, df.charges)
plt.title('Charges by Number of Children')
plt.xlabel('Number of Children')
plt.ylabel('Charges')
plt.show()
#len(y_test), len(reg_predict_test)
results = pd.DataFrame({'Actual':y_test, 'Predicted':reg_predict_test})
results
| Actual | Predicted | |
|---|---|---|
| 578 | 9724.53000 | 11392.0 |
| 610 | 8547.69130 | 9600.0 |
| 569 | 45702.02235 | 37696.0 |
| 1034 | 12950.07120 | 16384.0 |
| 198 | 9644.25250 | 6848.0 |
| ... | ... | ... |
| 651 | 10579.71100 | 14592.0 |
| 523 | 5397.61670 | 9600.0 |
| 495 | 1967.02270 | 2432.0 |
| 781 | 1146.79660 | 5760.0 |
| 581 | 1639.56310 | 3072.0 |
536 rows × 2 columns
sns.regplot(x = y_test, y = reg_predict_test, ci = None, color = 'b')
# sns.regplot(x = y_test, y = reg_predict_test, ci = None, scatter_kws = {'color': 'black'}, line_kws = {'color': 'red'})
<AxesSubplot:xlabel='charges'>
sns.lmplot(x='Actual', y='Predicted', data = results, fit_reg=False, height = 7)
# Plotting the diagonal line
line_coords = np.arange(results[['Actual', 'Predicted']].min().min()-10,
results[['Actual', 'Predicted']].max().max()+10)
plt.plot(line_coords, line_coords, # X and y points
color='darkorange', linestyle='--')
plt.ylabel('Predicted', fontsize=14)
plt.xlabel('Actual', fontsize=14)
plt.title('Linearity Assumption', fontsize=16)
plt.show()
fig, axes = plt.subplots(ncols = 3, figsize = (17,6), squeeze=True)
sns.scatterplot(x='bmi', y='charges', ax=axes[0], data=df,hue='sex')
sns.scatterplot(x='bmi', y='charges', ax=axes[1], data=df,hue='smoker')
sns.scatterplot(x='bmi', y='charges', ax=axes[2], data=df,hue='region')
<AxesSubplot:xlabel='bmi', ylabel='charges'>
os.system('jupyter nbconvert --to html "C:\Users\laram\OneDrive - Bellevue University\2022 Fall\DSC 450\Health_Insurance_Premiums.ipynb")
Input In [7] os.system('jupyter nbconvert --to html "C:\Users\laram\OneDrive - Bellevue University\2022 Fall\DSC 450\Health_Insurance_Premiums.ipynb') ^ SyntaxError: (unicode error) 'unicodeescape' codec can't decode bytes in position 31-32: truncated \UXXXXXXXX escape